Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A computerized method for generating dynamic look-alike data points of a seed dataset on a graphical user interface (GUI) comprising: defining a seed dataset from a collection of data, said collection of data comprising features associated with data; generating a first ranking of a first set of top features of the seed dataset; generating a second ranking of a second set of top features of the seed dataset; identifying an M×N set which is an M set of correlated features for each of a combined set N from the first set and the second set that are present in the seed dataset; for each one in the M×N set, generating a sliding scale cluster, in response to a user adjustment of a sliding scale slider displayed on the GUI while absent adjustments to the first set, the second set, the M set, the N set, and the M×N set, via permutation of the M sets of correlated features to be displayed on the GUI; generating a seed group meta bitmap index for a seed audience segment; generating a total audience bitmap index based on the generated seed group meta bitmap index; and for each sliding scale cluster, generating an available amplification count from the total audience bitmap index for display on the GUI in response to the user adjustment on the sliding scale slider while absent adjustments to the seed group meta bitmap index, and the total audience bitmap index, received from the GUI until a desired amplification count in the available amplification count is achieved.
This invention relates to a computerized method for generating dynamic look-alike data points of a seed dataset on a graphical user interface (GUI). The method addresses the challenge of identifying and visualizing similar data points within a larger dataset based on user-defined criteria. The process begins by defining a seed dataset from a collection of data, where the collection includes features associated with the data. A first ranking of top features from the seed dataset is generated, followed by a second ranking of another set of top features. An M×N set is then identified, which consists of M sets of correlated features for each of a combined set N from the first and second rankings that are present in the seed dataset. For each feature set in the M×N set, a sliding scale cluster is generated. The GUI displays a sliding scale slider that allows a user to adjust the clusters dynamically, triggering permutations of the M sets of correlated features without altering the original feature sets. The method also generates a seed group meta bitmap index for a seed audience segment and a total audience bitmap index based on the seed group meta bitmap index. For each sliding scale cluster, an available amplification count is derived from the total audience bitmap index and displayed on the GUI in response to user adjustments to the slider. This continues until the desired amplification count is achieved, all while maintaining the integrity of the seed group meta bitmap index and the total audience bitmap index. The system enables real-time exploration and refinement of look-alike data points based on user interactions.
2. The computerized method of claim 1 , wherein generating the ranking of the first set of top features of the seed dataset is based on an order of frequency of their occurrences.
The invention relates to a computerized method for analyzing datasets to identify and rank significant features. The method addresses the challenge of efficiently extracting meaningful patterns from large datasets by focusing on the most relevant features. The process begins with a seed dataset, which is a subset of the larger dataset containing initial data points. The method then identifies a first set of top features from the seed dataset, where these features are ranked based on their frequency of occurrence. Features that appear more frequently are given higher priority in the ranking. This frequency-based ranking helps in quickly identifying the most prominent features, which can then be used for further analysis or machine learning tasks. The method ensures that the most relevant and recurring features are prioritized, improving the efficiency and accuracy of subsequent data processing steps. By focusing on frequency, the approach reduces computational overhead and enhances the interpretability of the dataset. This technique is particularly useful in applications such as data mining, pattern recognition, and predictive modeling, where identifying key features is crucial for building effective models.
3. The computerized method of claim 1 , wherein generating the ranking of the second set of top features of the seed dataset is based on an order of frequency of their occurrence in correlation with their equivalent percentage in the entire collection of data.
This invention relates to a computerized method for ranking features in a dataset, particularly for identifying key features in a seed dataset by analyzing their frequency and distribution across a larger data collection. The method addresses the challenge of efficiently prioritizing features that are both frequent and representative of the broader dataset, ensuring meaningful insights while avoiding overemphasis on rare or isolated occurrences. The method involves generating a ranking of features from a seed dataset by evaluating their occurrence frequency in correlation with their proportional representation in the entire data collection. This dual-factor analysis ensures that features are ranked not just by how often they appear, but also by how consistently they appear relative to the overall dataset. For example, a feature that appears frequently in the seed dataset but is rare in the broader collection may be deprioritized, while a feature with moderate frequency but consistent distribution is ranked higher. The ranking process may also incorporate additional criteria, such as statistical significance or domain-specific relevance, to refine the results. The method is particularly useful in data mining, machine learning, and analytics, where identifying key features efficiently is critical for model training, pattern recognition, and decision-making. By balancing frequency and proportionality, the method provides a more robust and representative feature ranking than traditional approaches that rely solely on raw occurrence counts.
4. The computerized method of claim 1 , wherein generating the sliding scale cluster comprises sorting each permutation based on a closest occurrence match within the seed dataset.
This invention relates to data clustering techniques, specifically a method for generating a sliding scale cluster from a seed dataset. The problem addressed is the need for an efficient way to organize and analyze data permutations based on their relevance or similarity to existing data points. The method involves generating a sliding scale cluster by sorting each permutation within the dataset based on its closest occurrence match within the seed dataset. This ensures that permutations are grouped according to their proximity or similarity to known data points, improving the accuracy and usability of the clustering process. The sorting is performed using a computerized method, leveraging computational techniques to handle large datasets efficiently. The sliding scale cluster is then used to analyze or process the data further, depending on the application. This approach is particularly useful in fields such as data mining, pattern recognition, and machine learning, where organizing data based on similarity is crucial for extracting meaningful insights. The method ensures that permutations are ranked and grouped in a way that reflects their relevance to the seed dataset, enhancing the overall clustering performance.
5. The computerized method of claim 1 , further comprising displaying on the GUI each permutation as one cluster of data.
This invention relates to a computerized method for visualizing and analyzing permutations of data in a graphical user interface (GUI). The method addresses the challenge of presenting complex, multi-dimensional data permutations in a clear and organized manner, enabling users to better understand relationships and patterns within the data. The method involves generating permutations of data elements, where each permutation represents a unique combination of the elements. These permutations are then displayed on the GUI as distinct clusters, with each cluster visually representing one permutation. The clusters are arranged in a way that allows users to easily compare and analyze different permutations. The method may also include filtering or sorting the permutations based on user-defined criteria, such as relevance, frequency, or other metrics, to further refine the visualization. Additionally, the method may support interactive features, such as zooming, panning, or selecting individual clusters to view detailed information about the permutation. This interactive capability enhances usability by allowing users to explore the data in depth. The method may also include dynamic updates to the clusters in response to changes in the underlying data, ensuring that the visualization remains current. By organizing permutations into clusters and providing interactive visualization tools, this method improves data analysis workflows, particularly in fields such as bioinformatics, financial modeling, or machine learning, where understanding complex data relationships is critical.
6. The computerized method of claim 1 , further comprising identifying intersection counts across each cluster based on the seed map meta bitmap index to be displayed on the GUI.
This invention relates to data visualization and analysis, specifically for identifying and displaying intersection counts across clusters in a graphical user interface (GUI). The problem addressed is the need to efficiently analyze and visualize relationships between data clusters, particularly in large datasets where manual analysis is impractical. The method involves generating a seed map meta bitmap index, which is a data structure that organizes and indexes clusters of data points. This index allows for rapid identification of intersections between clusters, where an intersection refers to overlapping or shared data points between two or more clusters. The system then calculates intersection counts for each cluster, representing the number of shared data points between clusters. These counts are displayed on the GUI, providing users with a visual representation of cluster relationships. The GUI may include visual elements such as graphs, charts, or heatmaps to depict the intersection counts, enabling users to quickly identify highly interconnected clusters or outliers. The method enhances data analysis by automating the detection of cluster intersections and presenting the results in an intuitive format, improving decision-making in fields such as bioinformatics, market research, or network analysis. The invention streamlines the process of understanding complex datasets by leveraging computational techniques to highlight key relationships.
7. The computerized method of claim 1 , further comprising displaying each of the clusters with different colors or different shading of colors.
A computerized method for visualizing data clusters involves grouping data points into clusters based on similarity metrics and displaying each cluster with distinct visual identifiers. The method first processes input data to identify patterns or relationships, then applies clustering algorithms to categorize the data into distinct groups. Each cluster is assigned a unique visual representation, such as different colors or varying shades of a single color, to enhance visual differentiation. This approach improves data interpretation by making it easier for users to distinguish between clusters, particularly in large or complex datasets. The method may also include additional steps such as adjusting cluster boundaries or refining visual representations based on user feedback or predefined criteria. The use of distinct visual identifiers ensures clarity and reduces ambiguity in data analysis, making it suitable for applications in data mining, machine learning, and decision support systems. The method is designed to work with various types of data, including numerical, categorical, and textual information, and can be integrated into existing data visualization tools.
8. A computerized system for generating dynamic look-alike data points of seed dataset on a graphical user interface (GUI) comprising: a database system for storing a collection of data, said collection of data comprising features associated with data; wherein the database system defines a seed dataset from a collection of data in response to a user instruction; a processor configured to execute computer-executable instructions to generate a first ranking of a first set of top features of the seed dataset; wherein the processor generates a second ranking of a second set of top features of the seed dataset; wherein the processor identifies an M×N set which is an M set of correlated features for each of a combined set N from the first set and the second set that are present in the seed dataset; for each one in the M×N set, wherein the processor generates a sliding scale cluster, in response to a user adjustment of a sliding scale slider displayed on the GUI while absent adjustments to the first set, the second set, the M set, the N set, and the M×N set, via permutation of the M set of correlated features to be displayed on the GUI; wherein the processor generates a seed group meta bitmap index for a seed audience segment; wherein the processor generates a total audience bitmap index based on the generated seed group meta bitmap index; for each cluster, wherein the processor generates an available amplification count from the total audience bitmap index; and a display for displaying the available amplification count on the GUI, in response to the user adjustment on the sliding scale slider while absent adjustments to the seed group meta bitmap index, and the total audience bitmap index, received from the GUI, until a desired amplification count in the available amplification count is achieved.
The system is designed for generating dynamic look-alike data points from a seed dataset within a graphical user interface (GUI). It operates in the domain of data analysis and audience segmentation, addressing the challenge of identifying and expanding target audiences based on specific characteristics. The system includes a database storing a collection of data with associated features. A user can define a seed dataset from this collection. A processor then generates two rankings of top features from the seed dataset. It identifies an M×N set of correlated features by combining M sets of correlated features from each of N sets derived from the two rankings. For each feature set in the M×N combination, the processor creates a sliding scale cluster. A GUI slider allows the user to adjust the cluster dynamically by permuting the correlated features without altering the original feature sets or their combinations. The system also generates a seed group meta bitmap index for a seed audience segment and a total audience bitmap index based on this meta index. For each cluster, the processor calculates an available amplification count from the total audience bitmap index. The GUI displays this count in real-time as the user adjusts the slider, enabling iterative refinement until the desired amplification count is achieved. This approach facilitates precise audience expansion while maintaining feature correlations.
9. The computerized system of claim 8 , wherein the processor generates the first ranking of the first set of top features of the seed dataset based on an order of frequency of their occurrences.
The invention relates to a computerized system for analyzing datasets to identify and rank significant features. The system addresses the challenge of efficiently extracting meaningful patterns from large datasets by automating the identification and ranking of key features. The system processes a seed dataset to generate a first set of top features, which are then ranked based on their frequency of occurrence. This ranking helps prioritize features that appear most frequently, indicating their potential importance or relevance within the dataset. The system may also include additional components for further analysis, such as generating a second set of top features from a second dataset and comparing the two sets to identify common or unique features. The ranking process ensures that the most relevant features are highlighted, aiding in data interpretation and decision-making. The system is designed to handle large datasets efficiently, providing users with actionable insights derived from the ranked features.
10. The computerized system of claim 8 , wherein the processor generates the second ranking of the second set of top features of the seed dataset based on an order of frequency of their occurrence in correlation with their equivalent percentage in entire collection of data.
The system is designed for analyzing large datasets to identify and rank significant features. The problem addressed is the need to efficiently extract and prioritize relevant features from a dataset, particularly when dealing with high-dimensional data where manual analysis is impractical. The system processes a seed dataset to generate a first ranking of top features based on their frequency of occurrence. It then refines this ranking by generating a second ranking of the top features, where the ranking is determined by both the frequency of occurrence and their proportional representation within the entire dataset. This dual-factor approach ensures that features are evaluated not only by how often they appear but also by their relative significance across the full data collection. The system may also include a user interface for visualizing the ranked features, allowing users to explore and interpret the results. The method involves statistical analysis to correlate feature frequency with their distribution in the dataset, providing a more nuanced understanding of feature importance. This approach is particularly useful in fields like data mining, machine learning, and big data analytics, where identifying key features is critical for model training and decision-making.
11. The computerized system of claim 8 , wherein generating the sliding scale comprises sorting each permutation based on a closest occurrence match within the seed dataset.
The invention relates to a computerized system for generating a sliding scale from a seed dataset, addressing the challenge of efficiently organizing and presenting permutations of data points based on relevance or similarity. The system processes a seed dataset containing multiple data points and generates permutations of these points. Each permutation is then sorted based on a closest occurrence match within the seed dataset, meaning the permutations are ranked according to how closely they align with the original data points. This sorting ensures that the most relevant or similar permutations are prioritized, improving the usability and accuracy of the sliding scale. The system may also include additional features such as filtering permutations based on predefined criteria or adjusting the scale dynamically to reflect changes in the dataset. The overall goal is to provide a structured and intuitive way to analyze and compare permutations, enhancing decision-making processes in applications like data analysis, machine learning, or optimization tasks.
12. The computerized system of claim 8 , wherein the display displays on the GUI each permutation as one cluster of data.
A computerized system is designed to process and visualize data permutations in a graphical user interface (GUI). The system addresses the challenge of managing and presenting large sets of data permutations, which can be complex and difficult to interpret. The system organizes permutations into distinct clusters, each representing a unique permutation of the data. This clustering approach enhances readability and allows users to quickly identify and analyze different permutations without overwhelming them with raw, unstructured data. The GUI displays each permutation as a single cluster, making it easier to compare and evaluate variations. The system may also include additional features, such as filtering or sorting tools, to further refine the displayed permutations based on user preferences or specific criteria. By grouping permutations into clusters, the system simplifies the analysis of complex datasets, improving efficiency and decision-making. The technology is particularly useful in fields requiring permutation analysis, such as combinatorial optimization, data modeling, or algorithmic design.
13. The computerized system of claim 8 , wherein the processor identifies intersection counts across each cluster based on the seed map meta bitmap index to be displayed on the GUI.
This invention relates to a computerized system for analyzing and visualizing data clusters, particularly in the context of seed map meta bitmap indexing. The system addresses the challenge of efficiently identifying and displaying intersection counts across multiple data clusters in a graphical user interface (GUI). The processor within the system processes a seed map meta bitmap index, which organizes data into clusters and tracks their intersections. By analyzing this index, the processor calculates the number of intersections (overlaps) between each cluster, providing a quantitative measure of how these clusters relate to one another. These intersection counts are then displayed on the GUI, allowing users to visually assess cluster relationships and overlaps. The system may also include additional features such as generating the seed map meta bitmap index, clustering data, and enabling user interaction with the displayed clusters. The primary benefit is the ability to quickly and intuitively understand complex data relationships through visual representation of cluster intersections.
14. The computerized system of claim 8 , wherein the display displays each of the clusters with different colors or different shading of colors.
A computerized system for visualizing data clusters in a graphical user interface (GUI) is designed to enhance user comprehension of grouped data points. The system processes input data to identify and group similar data points into distinct clusters using clustering algorithms. Each cluster is then visually represented in the GUI with unique colors or varying shades of colors to distinguish them from one another. This color differentiation allows users to quickly identify and analyze the relationships between different clusters. The system may also include interactive features, such as zooming or filtering, to further refine the visualization. By providing clear visual distinctions between clusters, the system improves the efficiency of data analysis and decision-making processes. The technology is particularly useful in fields like data science, business intelligence, and machine learning, where understanding data patterns is critical. The use of color or shading ensures that users can easily interpret the clustering results without requiring extensive technical knowledge.
15. A computerized system for generating dynamic look-alike data points of a seed dataset on a graphical user interface (GUI) comprising: a database system for storing a collection of data, said collection of data comprising features associated with data; wherein the database system defines a seed dataset from a collection of data in response to a user instruction; a processor configured to execute computer-executable instructions to generate a first ranking of a first set of top features of the seed dataset; wherein the processor generates a second ranking of a second set of top features of the seed dataset; wherein the processor identifies an M×N set which is an M set of correlated features for each of a combined set N from the first set and the second set that are present in the seed dataset; for each one in the M×N set, wherein the processor generates a sliding scale cluster, in response to a user adjustment of a sliding scale slider displayed on the GUI while absent adjustments to the first set, the second set, the M set, the N set, and the M×N set, via permutation of the M set of correlated features to be displayed on the GUI, wherein the GUI defines a first data field for displaying data values for the sliding scale cluster and wherein the GUI defines a second graphical display area for graphically representing the sliding scale cluster; wherein the processor generates a seed group meta bitmap index for a seed audience segment; wherein the processor generates a total audience bitmap index based on the generated seed group meta bitmap index; for each cluster, wherein the processor generates an available amplification count from the total audience bitmap index; and a display for displaying the available amplification count on the GUI, in response to the user adjustment on the sliding scale slider while absent adjustments to the seed group meta bitmap index, and the total audience bitmap index, received from the GUI, until a desired amplification count in the available amplification count is achieved, wherein the display defines a graphical element and a corresponding text element of the available amplification count on the GUI.
The invention relates to a computerized system for generating dynamic look-alike data points from a seed dataset, enabling users to refine and visualize data clusters interactively. The system addresses the challenge of identifying and analyzing similar data points within large datasets by providing a user-adjustable interface for feature ranking and clustering. The system includes a database storing a collection of data with associated features. A seed dataset is defined from this collection based on user input. A processor ranks features from the seed dataset into two sets, then identifies correlated feature combinations (M×N) from these rankings. Users adjust a sliding scale on the GUI to dynamically permute and display clusters of correlated features without altering the underlying feature sets. The GUI includes fields for displaying data values and graphical representations of these clusters. The system also generates a seed group meta bitmap index for a specific audience segment and a total audience bitmap index. For each cluster, an available amplification count is derived from the total audience index, reflecting the potential reach or expansion of the seed dataset. This count is displayed on the GUI as both a graphical and text element, allowing users to refine the cluster until the desired amplification is achieved. The interactive adjustments enable real-time exploration of data similarities without modifying the core feature rankings or indices.
16. The computerized system of claim 15 , wherein the processor generates the first ranking of the first set of top features of the seed dataset based on an order of frequency of their occurrences.
This invention relates to a computerized system for analyzing datasets to identify and rank significant features. The system addresses the challenge of efficiently extracting meaningful patterns from large datasets, particularly when the data is unstructured or lacks clear hierarchical organization. The system processes a seed dataset to generate a ranked list of top features based on their frequency of occurrence, enabling users to quickly identify the most relevant or recurring elements within the data. The system includes a processor that analyzes the seed dataset to determine the frequency of each feature. Features are defined as distinct elements or attributes within the dataset, such as keywords, numerical values, or categorical labels. The processor then generates a first ranking of the top features by sorting them according to their occurrence frequency, with the most frequently occurring features appearing at the top of the list. This ranking helps users prioritize features that are most prevalent in the dataset, which can be useful for further analysis, decision-making, or data-driven insights. The system may also include additional components, such as a display for visualizing the ranked features or an interface for user interaction. The ranked output can be used to guide subsequent data processing steps, such as filtering, clustering, or predictive modeling. By focusing on frequency-based ranking, the system provides a straightforward yet effective method for feature selection in datasets where relevance is correlated with occurrence frequency. This approach is particularly valuable in applications like text mining, bioinformatics, or market research, where identifying dominant patterns is critical.
17. The computerized system of claim 15 , wherein the processor generates the second ranking of the second set of top features of the seed dataset based on an order of frequency of their occurrence in correlation with their equivalent percentage in entire collection of data.
The invention relates to a computerized system for analyzing and ranking features within a dataset. The system addresses the challenge of identifying and prioritizing the most relevant features from a large dataset, particularly when working with unstructured or high-dimensional data. The system processes a seed dataset to extract a set of top features, then generates a second ranking of these features based on their frequency of occurrence and their relative percentage within the entire data collection. This dual-ranking approach helps distinguish between frequently occurring features and those that are proportionally significant across the dataset. The system may also include a user interface for visualizing the ranked features and their relationships, aiding in data exploration and decision-making. The invention is particularly useful in fields like data mining, machine learning, and analytics, where feature selection and prioritization are critical for model performance and interpretability. The system dynamically adjusts rankings based on new data inputs, ensuring the relevance of the extracted features over time.
18. The computerized system of claim 15 , wherein generating the sliding scale comprises sorting each permutation based on a closest occurrence match within the seed dataset.
The system relates to data processing and analysis, specifically for generating a sliding scale from a seed dataset. The problem addressed is efficiently organizing permutations of data points to reflect their relevance or similarity to a reference dataset. The system sorts permutations based on the closest occurrence match within the seed dataset, ensuring that the most relevant permutations are prioritized. This involves comparing each permutation to the seed dataset to determine the nearest match, then ranking them accordingly. The sliding scale is dynamically generated to reflect these rankings, allowing for adaptive data analysis. The system may also include preprocessing steps to prepare the seed dataset, such as filtering or normalizing data, and may apply machine learning techniques to refine the matching process. The goal is to provide a structured, prioritized view of permutations that aligns with the seed dataset's characteristics, improving decision-making or pattern recognition in applications like recommendation systems, anomaly detection, or predictive modeling.
19. The computerized system of claim 15 , wherein the display displays on the GUI each permutation as one cluster of data.
A computerized system is designed to process and visualize complex datasets, particularly for applications requiring the analysis of multiple permutations or combinations of data elements. The system addresses the challenge of presenting large, multi-dimensional datasets in a way that is intuitive and actionable, often encountered in fields such as data science, bioinformatics, or financial modeling. The system includes a graphical user interface (GUI) that organizes and displays data in a structured manner, allowing users to interact with and interpret the information efficiently. The system generates and processes permutations of data elements, which may involve combinations of variables, parameters, or other data points. These permutations are then presented on the GUI as distinct clusters, where each cluster represents a unique permutation. This clustering approach helps users quickly identify and compare different permutations, reducing cognitive load and improving decision-making. The system may also include additional features such as filtering, sorting, or highlighting specific clusters based on user-defined criteria, further enhancing usability. The display of permutations as clusters allows for a more organized and visually coherent representation of the data, making it easier to analyze relationships, patterns, or outliers within the dataset. This approach is particularly useful when dealing with high-dimensional data or when the number of permutations is too large for traditional tabular or linear displays. The system may also support dynamic updates, allowing users to adjust parameters and see real-time changes in the displayed clusters.
20. The computerized system of claim 15 , wherein the processor identifies intersection counts across each cluster based on the seed map meta bitmap index to be displayed on the GUI.
This invention relates to a computerized system for analyzing and visualizing data clusters, particularly in the context of identifying and displaying intersection counts between clusters. The system addresses the challenge of efficiently processing and presenting complex data relationships in a user-friendly graphical interface. The system includes a processor that generates a seed map meta bitmap index, which serves as a compact representation of data clusters and their intersections. The processor then identifies intersection counts across each cluster using this index. These intersection counts quantify the degree of overlap or shared elements between different clusters, providing insights into data relationships. The results are displayed on a graphical user interface (GUI), allowing users to visualize and interpret the intersection data. The system may also include a memory for storing the seed map meta bitmap index and other relevant data, as well as a display for rendering the GUI. The GUI may feature interactive elements, such as zoom or filter controls, to enhance user exploration of the intersection data. The processor may further apply clustering algorithms to organize data into meaningful groups before calculating intersection counts. By leveraging the seed map meta bitmap index, the system efficiently computes intersection counts without exhaustive pairwise comparisons, improving performance and scalability. This approach is particularly useful in applications like bioinformatics, network analysis, or any domain requiring the analysis of large, interconnected datasets. The visualization of intersection counts helps users quickly identify key relationships and patterns within the data.
Unknown
September 17, 2019
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.